© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 7820
    Prev
    Next
    INDEX
    Explanations

    shapes, particularly geometric and curved forms in biological or physical contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    relating to shapes

    np_acts-logits-general · gemini-2.0-flash

    shape descriptors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    },[])
    -0.56
     Blind
    -0.52
    liferay
    -0.48
    Blind
    -0.47
     noDo
    -0.47
    blind
    -0.45
    serv
    -0.45
    ocere
    -0.44
    Backdrop
    -0.43
    viewDidLoad
    -0.43
    POSITIVE LOGITS
     shape
    2.15
     Shape
    1.86
     shaped
    1.83
     shapes
    1.81
    shape
    1.80
     SHAPE
    1.65
    Shape
    1.65
     Shaped
    1.62
    shaped
    1.57
     Shapes
    1.52
    Activations Density 0.460%

    No Known Activations